mbox series

[net-next,0/7] sfc: basic conntrack offload

Message ID cover.1691063675.git.ecree.xilinx@gmail.com (mailing list archive)
Headers show
Series sfc: basic conntrack offload | expand

Message

edward.cree@amd.com Aug. 3, 2023, 11:56 a.m. UTC 
From: Edward Cree <ecree.xilinx@gmail.com>

Support offloading tracked connections and matching against them in
 TC chains on the PF and on representors.
Later patch serieses will add NAT and conntrack-on-tunnel-netdevs;
 keep it simple for now.

Edward Cree (7):
  sfc: add MAE table machinery for conntrack table
  sfc: functions to register for conntrack zone offload
  sfc: functions to insert/remove conntrack entries to MAE hardware
  sfc: offload conntrack flow entries (match only) from CT zones
  sfc: handle non-zero chain_index on TC rules
  sfc: conntrack state matches in TC rules
  sfc: offload left-hand side rules for conntrack

 drivers/net/ethernet/sfc/Makefile       |   2 +-
 drivers/net/ethernet/sfc/bitfield.h     |   2 +
 drivers/net/ethernet/sfc/mae.c          | 827 +++++++++++++++++++++++-
 drivers/net/ethernet/sfc/mae.h          |  12 +
 drivers/net/ethernet/sfc/mcdi.h         |  14 +
 drivers/net/ethernet/sfc/tc.c           | 532 ++++++++++++++-
 drivers/net/ethernet/sfc/tc.h           |  86 ++-
 drivers/net/ethernet/sfc/tc_conntrack.c | 533 +++++++++++++++
 drivers/net/ethernet/sfc/tc_conntrack.h |  55 ++
 drivers/net/ethernet/sfc/tc_counters.c  |   8 +-
 drivers/net/ethernet/sfc/tc_counters.h  |   4 +
 11 files changed, 2038 insertions(+), 37 deletions(-)
 create mode 100644 drivers/net/ethernet/sfc/tc_conntrack.c
 create mode 100644 drivers/net/ethernet/sfc/tc_conntrack.h

Comments

Simon Horman Aug. 4, 2023, 1:44 p.m. UTC | #1 
On Thu, Aug 03, 2023 at 12:56:16PM +0100, edward.cree@amd.com wrote:
> From: Edward Cree <ecree.xilinx@gmail.com>
> 
> Support offloading tracked connections and matching against them in
>  TC chains on the PF and on representors.
> Later patch serieses will add NAT and conntrack-on-tunnel-netdevs;
>  keep it simple for now.

Hi Ed,

I did provide a minr nit on patch 7/7, in response to that patch,
but overall this patchset looks good to me.

For the series,

Reviewed-by: Simon Horman <horms@kernel.org>